Adjusting confirmed COVID-19 case counts for testing volume

When assessing the relative prevalence of the novel coronavirus (COVID-19), observers often point to the number of COVID-19 cases that have been confirmed through viral testing. However, comparisons based on confirmed case counts alone can be misleading since a higher case count may reflect either a higher disease prevalence or a better rate of disease detection. Using weekly records of viral test results for each state in the US, I demonstrate how confirmed case counts can be adjusted based on the percentage of COVID-19 tests that come back positive. A regression analysis indicates that case counts track better with future hospitalizations and deaths when employing this simple adjustment for testing coverage. Viral testing results can be used as a leading indicator of COVID-19 prevalence, but data reporting standards should be improved, and care should be taken to account for testing coverage when comparing confirmed case counts.